J. Ling et al. /Journal of Solid State Chemistry 178(2005)819-824 3 um 100um 10 um g (d)5/5.(e)6/4. where the inset image is a TEM image of single tetrapod. (f)7/3, and (g)9/1-see text for detale weight ratio of(b)2/8,(c) Fig. 2. SEM images of ZnO micro- and nanocrystals synthesized by (a) pure zinc and by Cu-zn all certain mechanisms: vapor-solid (VS)and catalyst-based therefore, Zn vapor can be released from the liquid mechanism Cu-Zn alloy slowly and continuously at the reaction In the cases of the pure Zn and the Zn mixed with a temperature. In addition, because Cuo can be reduced all or medium proportion of Cu, there is no catalyst to pure Cu by Zn vapors and then be re-oxidized again particles observed at the tips of arms of Zno tetrapod the CuO, coming from the oxidation of Cu on the structure, indicating that the growth of ZnO micro-and surface, also helped absorb a part of Zn vapor and nanostructure cannot be dominated by the catalyst depressed the releasing of Zn vapor, working well in based VLS mechanism but basically by a VS mechan- situ as an effective controller of metal gas. Previous ism. In fact, the additive Cu functions more likely a work on the growth of Zno gains proved a possible releasing controller of metal zinc gas rather than a shape control by adjusting the ratio of the oxygen catalyst as reported by other groups [8] partial pressure [3, 6]. Though some researches highlight According to the Cu-Zn binary phase diagram[9). the the importance of Zn vapor pressure, it is actually hard Cu addition increase the boiling point of Cu-Zn alloy, to search for a simple and effective method to controlmechanisms: vapor–solid (VS) and catalyst-based vapor–liquid–solid (VLS) mechanism. In the cases of the pure Zn and the Zn mixed with a small or medium proportion of Cu, there is no catalyst particles observed at the tips of arms of ZnO tetrapod structure, indicating that the growth of ZnO micro- and nanostructure cannot be dominated by the catalyst￾based VLS mechanism but basically by a VS mechan￾ism. In fact, the additive Cu functions more likely a releasing controller of metal zinc gas rather than a catalyst as reported by other groups [8]. According to the Cu–Zn binary phase diagram [9], the Cu addition increase the boiling point of Cu–Zn alloy, therefore, Zn vapor can be released from the liquid Cu–Zn alloy slowly and continuously at the reaction temperature. In addition, because CuO can be reduced to pure Cu by Zn vapors and then be re-oxidized again, the CuO, coming from the oxidation of Cu on the surface, also helped absorb a part of Zn vapor and depressed the releasing of Zn vapor, working well in situ as an effective controller of metal gas. Previous work on the growth of ZnO gains proved a possible shape control by adjusting the ratio of the oxygen partial pressure [3,6]. Though some researches highlight the importance of Zn vapor pressure, it is actually hard to search for a simple and effective method to control ARTICLE IN PRESS Fig. 2. SEM images of ZnO micro- and nanocrystals synthesized by (a) pure zinc and by Cu–Zn alloy at certain weight ratio of (b) 2/8, (c) 4/6, (d) 5/5, (e) 6/4, where the inset image is a TEM image of single tetrapod, (f) 7/3, and (g) 9/1—see text for details. J. Ling et al. / Journal of Solid State Chemistry 178 (2005) 819–824 821